About the role
You will optimize inference engines and configurations for customer workloads, tune performance for throughput and latency targets, and guide post-training and fine-tuning pipelines into production. You will support strategic accounts, lead technical onboarding, and share field insights to improve product and model roadmaps.
Responsibilities
- Select, configure, and optimize inference engines for hardware, model architectures, and workload profiles
- Develop configuration updates and tune KV cache, speculative decoding, tensor parallelism, and quantization
- Drive RL training runs and guide LoRA, SFT, DPO, RLHF, and GRPO pipelines into production
- Serve as the primary technical contact for strategic accounts and optimize endpoint configurations
- Align customers on inference and post-training configurations during onboarding
- Surface field insights and contribute to product, model, feature, and research adoption
Requirements
- 5+ years in a technical role focused on inference systems, open-source LLM deployment, or post-training workflows
- Expert hands-on experience with inference engines including vLLM, TensorRT-LLM, or SGLang
- Knowledge of KV cache tuning, speculative decoding, tensor parallelism, pipeline parallelism, and quantization
- Hands-on experience with LoRA, SFT, DPO, RLHF, and GRPO pipelines
- Knowledge of open-source models and model selection
- Strong Python skills and comfort in production environments
Benefits
- Startup equity
- Health insurance
- Remote-work flexibility
Requirements
- 5+ years in a technical role focused on inference systems, open-source LLM deployment, or post-training workflows
- Expert hands-on experience with inference engines including vLLM, TensorRT-LLM, or SGLang
- Knowledge of KV cache tuning, speculative decoding, tensor parallelism, pipeline parallelism, and quantization
- Hands-on experience with LoRA, SFT, DPO, RLHF, and GRPO pipelines
- Knowledge of open-source models and model selection